video-description-oversight
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is an instructional document outlining a human-AI oversight workflow for video metadata quality control. It does not include any executable code, scripts, or tool invocations.
- [EXTERNAL_DOWNLOADS]: The skill references legitimate research papers and public repositories on platforms such as arXiv, GitHub, and HuggingFace for methodology and benchmarking purposes. These are standard academic and technical sources.
- [NO_CODE]: There are no scripts, binaries, or automated tasks associated with this skill; it functions as a set of behavioral guidelines for an AI agent.
Audit Metadata